google-chat-api

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime ingests outsider-authored free text from the incoming HTTP request body (e.g., templates/webhook-handler.ts reads message from await request.json() and injects it into cardsV2[].card.sections[].widgets[].textParagraph.text), which is then sent to the agent/LLM context as part of the workflow output.